DoCRL: Double Critic Deep Reinforcement Learning for Mapless Navigation of a Hybrid Aerial Underwater Vehicle with Medium Transition.
Ricardo B. GrandoJunior Costa de JesusVictor Augusto KichAlisson Henrique KollingRodrigo da Silva GuerraPaulo L. J. Drews-JrPublished in: CoRR (2023)
Keyphrases
- underwater vehicles
- reinforcement learning
- fault identification
- function approximation
- actor critic
- reinforcement learning algorithms
- temporal difference
- transition model
- computer simulation
- learning algorithm
- mathematical models
- step size
- aerial images
- robot navigation
- learning process
- multi agent
- learning problems
- machine learning
- hybrid learning
- model free
- transfer learning
- optimal policy
- function approximators
- supervised learning
- state space
- natural actor critic